301 patients with knee osteoarthritis were randomized to ten sessions over 2 weeks of either high intensity electroacupuncture, low intensity electroacupuncture, or sham electroacupuncture. Patients were assessed for response to therapy based on pain scores and functional scores after 1 and 2 weeks of treatment. Effects on pain scores, including visual analog scale and numeric rating scale, were significantly greater in patients who received active electroacupuncture compared to sham, with effect after 2 weeks greater in those who received high intensity electroacupuncture. Within the first week of treatment, scores for conditioned pain modulation did not significantly differ between the three groups, but by 2 weeks, improvement in the outcome was greater in the active electroacupuncture groups compared to the sham group; again, greatest effect was observed in the high intensity group.